The very high free energy requirement for transport of H+ from blood to gastric lumen is derived from aerobic metabolism, apparently via the cytochrome system. Previous evidence (largely spectrophotometric) suggests a hypothesis in which the H+ "pump" is not driven by ATP, but derives energy from a component of the cytochrome system in the plasma membrane. Three sets of observations suggest that the cytochrome oxidase for this cytochrome chain may have properties different from the mitochondrial cytochrome oxidase, which may allow separation and identification of this oxidase and the rest of the chain. First, although acid secretion is maximal at 0.9 atm 02, implying no hypoxic regions in the tissue, cytochrome a3 is not fully oxidized, as it should be given its very low Km for 02. Second, carbon monoxide at high CO/O2 ratios (hyperbaric) does not markedly inhibit acid secretion, as it should with the standard cytochrome oxidase. Finally, oxygen uptake of mucosal fragments is quite insensitive to inhibition by the terminal oxidase inhibitor azide, again implying a non-standard oxidase. The latter two facts suggest that the gastric mucosa is unique among inhibitor-insensitive systems in that it can utilize this pathway for purposes other than heat production. To permit these studies, modifications are proposed to the current multiwavelength spectrophotometer to enable its use at elevated pressures, and to permit spectra and kinetics from a single tissue sample while monitoring physiological parameters. With this instrument, identification of the components of the secretory chain should be possible, leading to their eventual isolation and characterization. Moreover, spectrophotometry will be possible in tissues known to be rate-limited by O2 at 0.9 atm, such as elasmobranch and mammalian mucosae. Increased knowledge of the mechanism of gastric acid secretion should suggest modifications to the current treatments for disorders of this process such as gastric ulcer.